<b>Does dwell time influence rankings? Untangling a stubborn myth.</b>

<b>Does dwell time influence rankings? Untangling a stubborn myth.</b>

"Long dwell time tells Google your page is good" is repeated endlessly. Google denies using analytics dwell time as a ranking factor. The truth is more interesting than either camp admits.

I can't see Google's internals, so I looked at it sideways: for 110 pages I correlated on-site engagement (time on page, scroll depth from analytics) with subsequent 3-month ranking changes, controlling for starting position.

— On-page time vs ranking change: correlation near zero (~0.05). No support for direct dwell-time ranking.
— But <i>pogo-sticking</i> proxies — fast returns to the SERP inferred from short sessions followed by no conversion — weakly tracked with <i>declines</i> in volatile niches only.
— The cleaner signal was content that satisfied intent fast (answer up top), which correlated with both engagement AND ranking — making engagement a <i>symptom</i>, not a cause.

My read: Google likely uses SERP-side interaction signals (clicks, result-switching) more than your-site dwell time. Optimizing analytics dwell time directly is chasing a symptom. Optimize for fast intent satisfaction; engagement follows, and so might the SERP signals you can't see.

Method note: analytics from consenting site owners; pogo-sticking inferred, not observed.

Confidence: low — heavy inference, no access to Google's actual signals.